JPH04321741A - Air-fuel ratio learning control device for internal combustion engine - Google Patents

Abstract

PURPOSE:To make convergence compatible with accuracy in air fuel ratio learning and suppress stepwise change of the correction level when the appropriate operation regions are changed over to each other. CONSTITUTION:A learning map where the whole operation range is divided into 16 ranges and a learning map where the whole operation range is divided into 256 ranges are provided. Learning of air fuel ratio learning correction factors KBLRC1 in the respective 16 ranges is made. Along with the advance of learning, shifting is made to learning of air fuel ratio learning correction factors KBLRC2 in the respective 256 ranges. After the steps (S51 to S57) for transferring the correction share by correction factors KBLRC1 to the side of correction factors KBLRC2 (S51-S57) are then made, interpolation calculation is made from the learning map for respective 256 ranges and correction factors KBLRC2 matching the operation conditions are read out (S59). The final learning correction factors KBLRC are thus set by the read out correction factors KBLRC2 and learning correction factors KBLRCphi applicable to all operation conditions and used for correcting the basic fuel injection quantity Tp.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an air-fuel ratio learning control device for an internal combustion engine, and more particularly, a device for correcting the amount of fuel supplied so that the air-fuel ratio of an intake air-fuel mixture in an automobile internal combustion engine matches a target air-fuel ratio. Regarding.

0002

[Prior Art] Conventionally, in an internal combustion engine equipped with an electronically controlled fuel injection device having an air-fuel ratio feedback correction control function, Japanese Patent Laid-Open No. 60-90944 and Japanese Patent Laid-Open No. 61-1
As disclosed in Japanese Patent No. 90142 and the like, there are some that employ learning control of the air-fuel ratio.

Air-fuel ratio feedback correction control uses an oxygen sensor installed in the engine exhaust system to determine whether the actual air-fuel ratio is rich or lean with respect to a target air-fuel ratio (for example, stoichiometric air-fuel ratio), and performs air-fuel ratio feedback correction based on the determination result. The coefficient LMD is set by proportional/integral control, etc., and the basic fuel injection amount Tp is calculated from engine operating state parameters (for example, intake air flow rate Q and engine rotational speed N) that are related to the amount of air taken into the engine. By correcting with the air-fuel ratio feedback correction coefficient LMD, the actual air-fuel ratio is feedback-controlled to the target air-fuel ratio.

[0004] Here, the deviation of the air-fuel ratio feedback correction coefficient LMD from the reference value (target convergence value) is learned for each of the plurality of operating regions and is determined as a learning correction coefficient KBLR.
C (air-fuel ratio learning correction value) is determined, and the basic fuel injection amount Tp is corrected by the learning correction coefficient KBLRC so that the base air-fuel ratio obtained without the correction coefficient LMD approximately matches the target air-fuel ratio, and the air-fuel ratio During feedback control, the fuel injection amount Ti is calculated by further correcting it using the correction coefficient LMD.

[0005] As a result, fuel correction can be performed in response to different correction requests for each operating condition, and the air-fuel ratio feedback correction coefficient LMD can be stabilized near the reference value, thereby improving air-fuel ratio controllability. By the way, the air-fuel ratio learning correction coefficient KBLRC for each operating region is set in order to respond to differences in air-fuel ratio correction requests due to differences in operating conditions as described above, so the operating region is divided as finely as possible. It is desirable to have them learn. However, the learning correction coefficient KBL is divided into finely divided operating regions, and the learning correction coefficient KBL is
If RC is trained, the learning opportunities in each driving area will decrease, the convergence of learning will deteriorate, and learned areas and unlearned areas will coexist, resulting in large gaps between driving areas. A difference in fuel ratio will occur.

[0006] Therefore, the present applicant has provided a plurality of learning maps with different numbers of divisions in the driving area, and among these plurality of learning maps, a learning map with a smaller number of divisions and a learning unit with a wider driving area is provided. We have previously proposed an air-fuel ratio learning control device that is configured to perform learning from a map, and as the learning progresses, to shift to learning a learning map with a larger number of segments and a narrower operating range of learning units (Japanese Patent Application No. 282883
(see issue).

According to such air-fuel ratio learning, learning convergence is ensured by learning for each large unit operating region at the initial stage of learning, and as learning progresses, learning for each unit operating region is performed in more detail. This enables learning to accurately respond to differences in correction requests due to differences in conditions.

[0008]

[Problem to be Solved by the Invention] However, as described above, by providing multiple learning maps and narrowing the area in which the air-fuel ratio is to be learned as the learning progresses, it is possible to perform air-fuel ratio learning in detail for each operating area. However, since the learning correction value is learned as a representative value within each divided driving area, the learning correction value may change in steps when the corresponding driving area changes on the learning map. Unavoidably, such stepwise changes in the learned correction value may cause air-fuel ratio fluctuations, resulting in deterioration of exhaust properties (see FIG. 15).

In other words, errors in the base air-fuel ratio occur due to various factors, but the base air-fuel ratio does not change stepwise in response to changes in operating conditions, and is originally caused by errors in the air-fuel ratio learning correction value. Requirements also do not change in steps. However, as mentioned above, the air-fuel ratio learning correction value is learned as a typical correction level within an operating range of a certain size, so when there is a change in operating conditions across the boundaries of the operating range, the actual This results in stepwise changes in the learning correction value that do not correspond to changes in operating conditions. The range of change in this stepwise learning correction value is
Although it can be made smaller by narrowing the learning area, it is not practical to divide the driving area into extremes even if learning is to gradually narrow the learning area, and it is difficult to adjust the correction level between areas on the learning map. The situation was such that it was unavoidable that a difference in level would occur.

The present invention has been made in view of the above problems, and while ensuring learning convergence and performing air-fuel ratio learning for each detailed operating region, it is also possible to It is an object of the present invention to avoid stepwise changes in actual correction, and to configure the air-fuel ratio learning correction used for correction to change smoothly and accurately in response to changes in operating conditions.

[0011]

[Means for Solving the Problems] Therefore, an air-fuel ratio learning control device for an internal combustion engine according to the present invention is constructed as shown in FIG. In FIG. 1, the engine operating condition detecting means detects the engine operating condition including at least an operating parameter related to the amount of air taken into the engine, and the basic fuel supply amount setting means detects the engine operating condition based on the detected engine operating condition. to set the basic fuel supply amount.

The air-fuel ratio feedback correction value setting means compares the air-fuel ratio of the engine intake air-fuel mixture detected by the air-fuel ratio detection means with a target air-fuel ratio, and adjusts the actual air-fuel ratio to approach the target air-fuel ratio. Set an air-fuel ratio feedback correction value to correct the basic fuel supply amount. On the other hand, the storage means rewritably stores an air-fuel ratio learning correction value for correcting the basic fuel supply amount for each operating region divided into a plurality of operating regions based on engine operating conditions, and the air-fuel ratio learning means has air-fuel ratio feedback. Learn the deviation of the correction value from the target convergence value,
The air-fuel ratio learning is performed by correcting and rewriting the air-fuel ratio learning correction value stored in the storage means corresponding to the corresponding operating region in a direction that reduces the deviation, and the range of the corresponding operating region is gradually narrowed as the learning progresses. I will do it.

[0013]The correction value reading means reads out the air-fuel ratio learning correction value corresponding to the actual operating condition from the storage means by performing an interpolation calculation using the narrowest operating region learned by the air-fuel ratio learning means as a unit. Accordingly, the fuel supply amount setting means sets the final fuel supply amount based on the read air-fuel ratio learning correction value, the basic fuel supply amount, and the air-fuel ratio feedback correction value.

Here, the fuel supply control means drives and controls the fuel supply means based on the set fuel supply amount.

[0015]

[Operation] According to the air-fuel ratio learning control device configured as described above, the air-fuel ratio learning means corrects and rewrites the air-fuel ratio learning correction value for the relevant operating region in the storage means as the learning progresses, and the air-fuel ratio learning means corrects and rewrites the air-fuel ratio learning correction value for the relevant operating region in the storage means. At the beginning of learning, air-fuel ratio learning is performed in units of a relatively large operating region to ensure the learning progress speed, and as learning progresses, the region in which the learning correction value is rewritten as the operating region is Since the width is narrowed in stages, it is possible to learn with high accuracy differences in correction requests due to differences in operating conditions.

On the other hand, when the air-fuel ratio learning correction value for each operating region is read out from the storage means and used for correcting the basic fuel supply amount, the narrowest operating region on the storage means is carried out after the learning has sufficiently progressed. Since the interpolation calculation is performed in units of , the air-fuel ratio learning correction value for each operating region, which is discrete data for operating conditions, can be smoothly changed in response to changes in operating conditions, and the basic fuel supply amount can be corrected. can be applied. Further, since the interpolation calculation is performed using the narrowest driving region as the learning region, the accuracy of interpolation can be ensured to the maximum.

[0017]

[Examples] Examples of the present invention will be described below. In FIG. 2 showing one embodiment, air is taken into an internal combustion engine 1 from an air cleaner 2 via an intake duct 3, a throttle valve 4, and an intake manifold 5. As shown in FIG. Each branch of the intake manifold 5 is provided with a fuel injection valve 6 as a fuel supply means for each cylinder. The fuel injection valve 6 is an electromagnetic fuel injection valve that opens when the solenoid is energized and closes when the energization is stopped, and opens when the solenoid is energized by a drive pulse signal from the control unit 12,
The engine 1 is intermittently injected with fuel that is pressure-fed from a fuel pump (not shown) and adjusted to a predetermined pressure by a pressure regulator.

Each combustion chamber of the engine 1 is provided with an ignition plug 7, which ignites a spark to ignite and burn the air-fuel mixture. Then, exhaust gas is discharged from the engine 1 via an exhaust manifold 8, an exhaust duct 9, a three-way catalyst 10, and a muffler 11. The control unit 12 includes a CPU,
Equipped with a microcomputer including ROM, RAM, A/D converter, input/output interface, etc.
It receives input signals from various sensors, performs arithmetic processing as described later, and controls the operation of the fuel injection valve 6.

The various sensors mentioned above include the intake duct 3
An air flow meter 13 is provided therein, and outputs a signal corresponding to the intake air flow rate Q of the engine 1. Further, a crank angle sensor 14 is provided, and in the case of the four-cylinder engine of this embodiment, outputs a reference signal REF for every 180 degrees of crank angle and a unit signal POS for every 1 degree or 2 degrees of crank angle. Here, the engine rotational speed N can be calculated by measuring the period of the reference signal REF or the number of occurrences of the unit signal POS within a predetermined period of time.

A water temperature sensor 15 is also provided to detect the temperature Tw of the cooling water in the water jacket of the engine 1. Here, the air flow meter 13, crank angle sensor 14, water temperature sensor 15, etc. correspond to the engine operating condition detection means in this embodiment, and the operating parameters related to the amount of air taken into the engine are These are the intake air flow rate Q and the engine rotation speed N.

Further, an oxygen sensor 16 as an air-fuel ratio detecting means is provided at the gathering part of the exhaust manifold 8, and detects the air-fuel ratio of the intake air-fuel mixture via the oxygen concentration in the exhaust gas. The oxygen sensor 16 detects that the oxygen concentration in the exhaust gas is at the stoichiometric air-fuel ratio (
This is a known method that detects whether the actual air-fuel ratio is rich or lean with respect to the stoichiometric air-fuel ratio by utilizing sudden changes after the target air-fuel ratio (in this example). A relatively high voltage signal is output when the air-fuel ratio is rich, and 0V when the air-fuel ratio is lean.
It shall output a nearby low voltage signal.

[0022] Here, the control unit 12
The CPU of the microcomputer built into the
Arithmetic processing is performed according to the programs on the ROM shown in the flowchart of FIG. 11, and the fuel injection amount Ti is set while executing air-fuel ratio feedback correction control and air-fuel ratio learning correction control for each operating region, and fuel is supplied to the engine 1. control.

In this embodiment, the functions as the basic fuel supply amount setting means, the fuel supply amount setting means, the fuel supply control means, the air-fuel ratio feedback correction value setting means, the air-fuel ratio learning means, and the correction value reading means are as follows. As shown in the flowcharts of FIGS. 3 to 11, the control unit 12
is provided in terms of software, and a RAM with a backup function of a microcomputer (not shown) built into the control unit 12 serves as storage means.

The program shown in the flowcharts of FIGS. 3 and 4 uses proportional and integral control to control the air-fuel ratio feedback correction coefficient LMD (air-fuel ratio feedback correction value), which is multiplied by the basic fuel injection amount (basic fuel supply amount) Tp. This is a program to set, and is executed every one revolution (1 rev) of the engine 1. First, in step 1 (indicated as S1 in the figure, the same applies hereinafter), the oxygen sensor (O2 /S) 16
Reads the voltage signal output according to the oxygen concentration in the exhaust gas.

[0025] In the next step 2, step 1
The voltage signal from the oxygen sensor 16 read in and the slice level (for example, 50
0 mV). When it is determined that the voltage signal from the oxygen sensor 16 is greater than the slice level and the air-fuel ratio is richer than the stoichiometric air-fuel ratio, the process proceeds to step 3, where it is determined whether or not the current rich determination is the first.

When the rich determination is made for the first time, the process proceeds to step 4, where the air-fuel ratio feedback correction coefficient LMD set up to the previous time is set to the maximum value a. In the next step 5, a predetermined proportionality constant P is subtracted from the previous correction coefficient LMD to control the correction coefficient LMD to decrease. Further, in step 6, a flag FP indicating that proportional control has been executed is set to 1.

On the other hand, if it is determined in step 3 that the rich determination is not the first time, the process proceeds to step 7, and the value obtained by multiplying the integral constant I by the latest fuel injection amount Ti is subtracted from the previous correction coefficient LMD. and updates the correction coefficient LMD. In addition, when it is determined in step 2 that the voltage signal from the oxygen sensor 16 is smaller than the slice level and that the air-fuel ratio is lean relative to the target, first, in step 8, the air-fuel ratio is determined to be lean relative to the target. It is determined whether or not the lean determination is the first time, and if it is the first time, the process advances to step 9 and the correction coefficient LMD up to the previous time is set to the minimum value b.

In the next step 10, the fuel injection amount Ti is increased by updating the previous correction coefficient LMD by adding the proportionality constant P, and in step 11, it is shown that the proportional control has been executed. Set flag FP to 1. When it is determined in step 8 that the lean determination is not the first time, the process proceeds to step 12, and the value obtained by multiplying the integral constant I by the latest fuel injection amount Ti is used as the correction coefficient L up to the previous time.
MD and gradually increases the correction coefficient LMD.

[0029] At the first time of rich/lean discrimination, the correction coefficient LM
When proportional control D is executed, various processes related to air-fuel ratio learning correction control, which will be described later, are further performed. In this embodiment, as shown in FIG. 12, the entire operating range is divided into 16 unit operating ranges, and the learning correction coefficient KBLRC1 is rewritably stored for each unit operating range.
The air-fuel ratio learning control device comprises a region learning map and a 256 region learning map that divides the entire operating region into 256 unit operating regions and stores the learning correction coefficient KBLRC2 for each unit operating region in an updatable manner. be done. Further, the learning correction coefficient KBLRC φ, which is applied under all operating conditions without dividing the operating range, is separately set for learning.

When proportional control of the air-fuel ratio feedback correction coefficient LMD is performed, first, in step 13, 1
A count value cnt is determined to determine whether the vehicle is stably stopped in one driving region on the six-region learning map. In the programs shown in the flowcharts of FIGS. 5 to 8, which will be described later, when the corresponding operating region on the 16-region learning map changes at predetermined minute intervals, the count value cnt is set to a predetermined value (for example, 4). The count value cn is set in step 13.
If it is determined that t is not zero, the process proceeds to step 14, where the count value cnt is decreased by 1.

In step 15, it is determined whether or not most of the driving areas on the 16 area learning map have been learned by checking a learned flag F[B, A, which is set for each area as described later.
]. When most of the 16 areas on the 16 area learning map have been learned, the process proceeds to step 16, where it is determined whether the corresponding driving area on the 256 area learning map is the same as the previous time. Then, the process proceeds to step 17 only when there is a change in the relevant operating range.

In step 17, the latest correction coefficient LMD
Target convergence value Target (=1
．． Based on the absolute value of the deviation with respect to 0), a map of Δstress indicating the degree of inappropriateness of the learned value is referred to, and Δstress is set to increase in accordance with an increase in the deviation of the correction coefficient LMD from the target convergence value Target. Then, the Δ stress obtained this time is added to the "stress" in which the integrated value of the Δ stress is set.

As will be described later, when the stress exceeds a predetermined value, the already learned air-fuel ratio learning correction coefficient KBLRC
It is determined that this is inappropriate and the student is forced to start over from the beginning. In other words, when learning has progressed sufficiently, the correction coefficient LMD should be stable at approximately the target convergence value Target even if the operating conditions change, but if the learning result is inappropriate, the correction coefficient LMD should be changed. Since we are trying to compensate for this, the target convergence value T of the correction coefficient LMD
The learning result is judged as good or bad according to the deviation with respect to arget, and if the learning result is inappropriate, the learning is restarted from the beginning so that an appropriate learning result can be obtained.

The programs shown in the flowcharts of FIGS. 5 to 8 are air-fuel ratio learning programs for each operating region, and are executed at predetermined minute intervals (for example, 10 ms). In step 21, the flag FP is determined, and FP is 1.
When the value is zero, the process proceeds to step 22, where the FP is reset to zero, and various processing by this program is performed, and when it is zero, the program is immediately terminated.

After the FP is reset to zero in step 22, in the next step 23, the learning correction coefficient KBLRC φ (initial value 1
．． A flag Fφ indicating whether or not 0) has been learned is determined. Here, if the flag Fφ is zero and learning of the learning correction coefficient KBLRCφ has not been completed, the process proceeds to step 24, and the maximum and minimum values a of the correction coefficient LMD are
, b is approximately 1.0.

When (a+b)/2 is not approximately 1.0, the process proceeds to step 26, and the target convergence value Target (in this embodiment, 1.0) of the correction coefficient LMD is determined from (a+b)/2.
) is multiplied by a predetermined coefficient X to the previous learning correction coefficient KBLRC φ, and the addition result is set as the new learning correction coefficient KBLRC φ. KB
LRC φ←KBLRC φ+X{(a+b)/2-T
Furthermore, in step 26, the learning correction coefficient KBLRC1 and the learning correction coefficient KBLRC2 stored in the respective driving regions of the 16-area learning map and the 256-area learning map are reset to the initial value of 1.0.

In step 24, (a+b)/2 is approximately 1.
If it is determined that the flag Fφ is
is set to 1 so that it is determined that learning of the learning correction coefficient KBLRCφ corresponding to all driving ranges has been completed. On the other hand, if it is determined in step 23 that the flag Fφ is 1, then the operating range is changed to the basic fuel injection amount Tp.
Air-fuel ratio learning is performed for each operating region divided into a plurality of regions based on the engine speed and the engine rotational speed N.

First, in step 27, on the 256 area learning map, the area [K, I
] is determined based on the basic fuel injection amount Tp and threshold values Tp[i] and N[i] of the engine rotational speed N, which are set in advance to divide the operating range into 256 ranges. here,
As shown in FIG. 12, K indicates the position of the corresponding region in 16 regions divided using the engine rotational speed N as a parameter, and I indicates the position of the corresponding region among 16 regions divided using the basic fuel injection amount Tp as a parameter.

In the next step 28, the corresponding area [K
,I] on the 16-area learning map,
Detect the corresponding area [B, A] on the 16-area learning map. Then, in the next step 29, it is determined whether the corresponding area on the 16-area learning map is the same as the previous one. Then, when the corresponding driving region changes on the 16-region learning map, the process proceeds to step 30, and the step 1
A predetermined value (for example, 4) is set to the count value cnt, which is decremented by 1 at 4.

In step 31, a flag F[B, A] indicating whether or not learning in area [B, A] has been completed in the 16 area learning map is determined, and this flag F[
B, A] is zero and learning in the area [B, A] is not completed, the process advances to step 32. Step 3
In step 2, it is determined whether or not the count value cnt is zero, and if the count value cnt is not zero and there is a change in the corresponding area in the 16-area learning map, the program is immediately terminated and the count value cnt is determined to be zero. The process proceeds to step 33 only when the corresponding driving area is stable on the 16-area learning map.

In step 33, the maximum and minimum values a of the air-fuel ratio feedback correction coefficient LMD sampled by the program shown in the flowcharts of FIGS. 3 and 4 are determined.
, b, that is, the center value of the correction coefficient LMD is the initial value (=1
．． 0) Determine the progress of learning based on whether or not it is in the vicinity. Here, if it is not recognized that the average value of the correction coefficient LMD is approximately 1.0 and learning has not been completed, the process directly proceeds to step 35, and the average value of the correction coefficient LMD is approximately 1.0.
If it is determined that the learning has been completed, the flag F[B, A] is set to 1 in step 34, and the process proceeds to step 35.

In step 35, a target convergence value Target is calculated from the average value of the maximum and minimum values a and b for the learning correction coefficient KBLRC1 stored corresponding to the area [B, A] on the 16-area learning map. (1.0 in this example) is multiplied by a predetermined coefficient X1, and the result is added to the current driving area [B, A] on the 16-area learning map.
A new learning correction coefficient KBLRC1 is set corresponding to , and the map data is updated.

[0043]KBLRC1[B,A]←KBLRC1[
B, A] +
B, A] Learning correction coefficient KBL of 16 areas included in area
For RC2, in step 36, all these are set to the initial value 1.
．． Reset to 0. On the other hand, when it is determined in step 31 that the flag F [B, A] is 1, the process proceeds to step 37, and the driving region [B, A] on the 16 region learning map is determined to be 1.
A] is further subdivided into 16 regions.256 Region learning map learning is then started.

In step 37, it is determined whether (a+b)/2, which is the average value of the correction coefficient LMD, approximately matches the target convergence value Target of 1.0, and (a+b)/2 is determined.
b) If /2 is not approximately 1.0 and is in an unlearned state requiring correction by the air-fuel ratio feedback correction coefficient LMD, the process proceeds to step 38. In step 38, (
The value obtained by subtracting the target convergence value Target (1.0 in this example) from a+b)/2 and multiplying by a predetermined coefficient X2 is calculated as the driving region [K, I ] is added to the learning correction coefficient KBLRC2 [K, I] stored corresponding to
Set as RC2 [K, I] and update the map data.

[0045] KBLRC2 [K, I]←KBLRC2
[K, I]+X2 {(a+b)/2-Target
}Meanwhile, in step 37, (a+b)/2, which is the average value of the correction coefficient LMD, is 1.0 of the target convergence value Target.
If it is determined that they substantially match, step 39
Step 256: Flag FF [K, I] is set to 1 so that it is determined that the learning of the driving region [K, I] that includes the current driving conditions of the region learning map is completed.

[0046] From step 40 onwards, for the predetermined driving area [K, I] on the 256 area learning map for which it is determined that the learning has been completed this time, multiple driving areas with similar driving conditions are adjacent to each other on the same map. For each, control is performed to set an appropriate learning correction coefficient KBLRC2 estimated based on the learning correction coefficient KBLRC2 in the corresponding region [K, I]. In step 40, m and n are set to values obtained by subtracting 1 from K and I, respectively, which indicate the area position where the current operating condition is included in the 256 area learning map, and in the next step 41, it is determined whether m=K+2 or not. Determine whether

When the process proceeds from step 40 to step 41, a NO determination is made in step 41, so the process proceeds to step 42, where learning of the driving area on the 256 area learning map indicated by [m, n] is completed. whether or not
The determination is made based on whether the flag FF [m, n] is 1 or zero. Here, if the flag FF [m, n] is zero and learning has not been completed, the process proceeds to step 43. In step 43, the area [m, n] is converted from the area position [m, n] on the 256 area learning map to the area position [m/4, n/4] on the 16 area learning map. Identify the areas included on the 16-area learning map. Then, the area [m, n] adjacent to the corresponding area [K, I] on the 256 area learning map is 1
It is determined whether or not it is included in the same area [B, A] as [K, I] on the 6-area learning map.

That is, [K, I] is a region included in [B, A], but the adjacent region of [K, I] is another region adjacent to [B, A] on the 16-region learning map. This is because it may be included in the area, and the corresponding area [K, I] and the adjacent area [
m, n] are included in the same [B, A], then ([m
/4,n/4]=[B,A]), proceed to step 44,
The learning correction coefficient KBLRC2 corresponding to the [K, I] region that was determined to have been learned this time is directly applied to the adjacent region [m,
n] is stored as a learned value.

On the other hand, in step 43, the corresponding area [K, I]
When the adjacent area [m, n] is determined to be included in a different area on the 16 area learning map, ([m/4, n
/4]≠[B,A]), the process proceeds to step 45, and the learning correction coefficient KBL calculated by the following formula is added to the adjacent region [m, n].
Store RC2. KBLRC2 [m, n]←KBL
RC1 [B, A] + KBLRC2 [K, I] - KBLR
C1 [m/4, n/4] KBLRC2 as above
When [m, n] is updated, in step 46, m is increased by 1 and the process returns to step 41, where m=K+
It is determined whether the learning has been completed or not for each driving region until it reaches 2.

Then, as a result of the 1-up processing of m in step 46, when it is determined in step 41 that m=K+2, the process proceeds to step 47, and it is determined whether or not n=I+2, and n≠I+2. , step 4
In step 8, m is set to K-1 again, and in the next step 49, n is incremented by 1, and then the process proceeds to step 42. When it is determined in step 47 that n=I+2, [K,
This means that the determination processing for all eight operating regions surrounding [I] has been completed, so at this time, proceed to step 38,
The learning correction coefficient KBLRC2, which is determined to have already been learned in the current area [K, I], is updated.

As described above, in this embodiment, first, the learning correction coefficient KBLRC φ corresponding to all driving regions is learned, and then learning is performed for each driving region on the 16 region learning map. For areas that have already been trained on the 16-area learning map, that area is further divided into 16 areas for learning, so learning is performed in stages from large driving areas to small driving areas. Learning proceeds by narrowing the size of the region, and learning for each large operating region ensures convergence of the air-fuel ratio, and
As the learning progresses, the learning is performed using a detailed operating region as the relevant operating region, so it is possible to respond accurately to differences in the required correction value due to differences in operating conditions.

The results learned as described above are shown in FIG.
The learning correction coefficient KBLRC corresponding to the actual driving conditions is read out from the learning results. To briefly describe the processing contents of the program shown in the flowchart of FIG. 9, the target convergence value Tar of the learning correction coefficient KBLRC1
By adding the deviation from get to the learning correction coefficient KBLRC2 of each of the 16 regions included in the driving region to which the learning correction coefficient KBLRC1 is applied,
The correction burden due to the learning correction coefficient KBLRC1 is transferred to the learning correction coefficient KBLRC2 side, and after this correction is transferred, all the learning correction coefficients KBLRC1 are reset to the target convergence value Target, and double Avoid air/fuel ratio correction. That is, 16 area learning maps and 256
The learning results for each area learning map are
The 256 areas on the 256 area learning map are summarized into learning correction coefficients KBLRC2 for each area.

The program shown in the flowchart of FIG. 9 is processed in the background, and first, in step 51, counters i and j for indicating each area on the 256 area learning map are reset to zero, and then Then, in step 52, it is determined whether the counter i has exceeded 15. When the counter i is 15 or less, the process proceeds to step 53, where the learning correction coefficient KBLRC2 [j, i] corresponding to the area on the 256 area learning map indicated by [j, i] is set to K2, Also,
Area [j on the 16-area learning map that includes [j, i]
/4, i/4] corresponding to the learning correction coefficient KBLRC1[
j/4, i/4] to K1.

Next, in step 54, the learning correction coefficient KBLRC2[j,i] is updated according to the following equation. K
BLRC2[j,i]←K2+(K1-Target)
In the above formula, 256 Learning correction coefficient K on the area learning map
If BLRC2[j,i] is updated, the learning correction coefficient KB
The correction burden due to LRC1 is the learning correction coefficient KBLRC.
This will be transferred to the second side.

After the learning correction coefficient KBLRC2 is updated in step 54, the counter i is incremented by 1 in step 55, and the process returns to step 52. Here, counter i is 15
When the counter j exceeds 15, the process proceeds from step 52 to step 56, where it is determined whether the counter j has exceeded 15 or not. Then, when counter j is 15 or less,
At step 57, the counter i is reset to zero, and
By incrementing the counter j by 1 and returning to step 52 again, step 5 is performed for all 256 areas.
The update in step 4 is performed.

256 When the update of the learning correction coefficient KBLRC2 on the area learning map is completed and it is determined in step 56 that the counter j has exceeded 15, the entire correction burden due to the learning correction coefficient KBLRC1 is replaced by the learning correction coefficient. KB
Since this has been transferred to the LRC2 side, in step 58, 1 on the 16 area learning map is
A process is performed to reset all the learning correction coefficients KBLRC1 corresponding to each of the six regions to the target convergence value Target (=1.0), that is, to the initial value.

The process of transferring the correction burden due to the learning correction coefficient KBLRC1 to the learning correction coefficient KBLRC2 side and resetting all the learning correction coefficients KBLRC1 to their initial values is completed, and the learning results are calculated for each learning map and for each driving region. 256 Learning correction coefficient KBLRC on the area learning map
2, in the next step 59, the 256
A learning correction coefficient KBLRC2 corresponding to the current driving condition is read out from the region learning map by performing linear interpolation.

Here, the learning correction coefficient KBLRC2 is the air-fuel ratio learning correction value learned for each of the narrowest operating regions in this embodiment, and as described above, the learning correction coefficient KBLRC2 on the 16-region learning map The burden of KBLRC1 is transferred to this, and the correction request level in each of the 256 areas is approximately faithfully represented. Therefore, there will not be a large difference in the correction level between each operating region, but if you perform linear interpolation as described above and read out data that corresponds to the actual operating conditions, learning correction will be made in response to changes in actual operating conditions. Since the coefficient KBLRC2 can be smoothly changed and the interpolation calculation is performed using the narrowest operating range as a unit, highly accurate interpolation calculation is possible. Therefore, even if the corresponding area on the learning map changes, the learning correction level will not change in steps.
Accurate learning correction coefficient KBLR according to the required correction level
Since C can be changed, air-fuel ratio fluctuations due to changes in operating conditions can be effectively suppressed.

In the interpolation calculation, the basic fuel injection amount Tp and the threshold values Tp[i] and N[i] of the engine rotational speed N, which separate the regions of the 256 region learning map, are used as shown in FIGS. 13 and 14. As shown, the center value (Tp[i]-
Tp[i-1])/2, (N[i]-N[i-1])/
2 is determined, and the center values RTp[i] and RN[i] are assumed to be the operating conditions to which the learning correction coefficient KBLRC2 in that region corresponds, and interpolation calculations are performed. This means that the learning correction coefficient KBLRC2 learned for each driving region is
Since it shows the average correction request in that area,
This is to further improve the accuracy of interpolation calculations, assuming that the data corresponds to average operating conditions within the region. The center value of the operating conditions in each region used for the interpolation calculation may be stored in advance on the ROM, or may be calculated based on the threshold value Tp[i], N[i] data on the ROM. .

In the above embodiment, the learning correction coefficient KBLRC
The learning result was corrected so that the learning correction coefficient KBLRC2 would bear the correction amount due to 1, but the learning correction coefficient KBLRC1 and the learning correction coefficient K as described above
256 without transferring the correction burden between BLRC2 and BLRC2.
In the interpolation calculation between regions on the region learning map, the result of adding the learning correction coefficient KBLRC1 corresponding to the data of the learning correction coefficient KBLRC2 may be used as discrete data.

Furthermore, as the interpolation calculation, linear interpolation, which is easy to calculate, is used in this embodiment, but if the calculation burden is acceptable, for example, cubic interpolation calculation may be used. Learning correction coefficient KBLRC2 corresponding to actual driving conditions
is read out from the map by linear interpolation, and in the next step 60, the final learning correction coefficient K is determined as follows.
Set up BLRC.

0062 KBLRC ←KBLRC φ+KBLR
The C2-Target learning correction coefficient KBLRC is a correction term that is multiplied by the basic fuel injection amount Tp, and each learning correction coefficient KBLRCφ, KBLRC2 changes its initial value to the target convergence value Targ of the air-fuel ratio feedback correction coefficient LMD.
et (=1.0), so the learning correction coefficient KBLRC φ corresponding to all driving conditions and the learning correction coefficient KBL corresponding to the corresponding area on the area learning map
From the added value with RC2, the target convergence value Target (=1
．． 0) has been subtracted. In addition, the learning correction coefficient KBLRC1 of the 16 area learning map is all the target convergence value Target.
Therefore, the learning correction coefficient KBLRC1 was not involved in the setting of the final learning correction coefficient KBLRC, but KBLRC ←KBLRC φ+KB
Learning correction coefficient KBLR including learning correction coefficient KBLRC1 as LRC1+KBLRC2-2×Target
C may be set.

The learning correction coefficient KBLRC finally set by the program shown in the flowchart of FIG. 9 above is as shown in FIG.
This is used in the fuel injection amount setting program shown in the flowchart of No. 0. The fuel injection amount setting program shown in the flowchart of FIG.
s), and first, step 81
Now, the engine rotation speed N calculated based on the intake air flow rate Q detected by the air flow meter 13 and the detection signal from the crank angle sensor 14 is input.

Then, in the next step 82, the basic fuel injection amount Tp (←K×Q /N; K is a constant). In the next step 83, various corrections are made to the basic fuel injection amount Tp calculated in step 82 to calculate the final fuel injection amount (fuel supply amount) Ti. Here, the correction values used to correct the basic fuel injection amount Tp are the learning correction coefficient KBLRC, the air-fuel ratio feedback correction coefficient L
Cooling water temperature T detected by MD and water temperature sensor 15
various correction coefficients COEF that are set including a basic correction coefficient based on w, a post-start increase correction coefficient, etc., and a correction amount Ts for correcting changes in the effective injection time of the fuel injection valve 6 due to changes in battery voltage. Yes, Ti←Tp×LMD×
The final fuel injection amount Ti is updated at predetermined time intervals by calculating KBLRC×COEF+Ts.

When a predetermined fuel injection timing is reached, the control unit 12 adjusts the latest calculated fuel injection amount T.
A drive pulse signal with a pulse width corresponding to i is sent to the fuel injector 6.
and controls the amount of fuel supplied to the engine 1. Further, the program shown in the flowchart of FIG. 11 is a process based on the "stress" (the integrated value of the deviation of the air-fuel ratio feedback correction coefficient LMD from the target convergence value) sampled according to the program shown in the flowcharts of FIGS. 3 and 4. This is a program that executes as a background job (BGJ).

In step 91, the parameters shown in FIGS. 3 and 4 are used as parameters indicating the degree of inappropriateness of the air-fuel ratio learning correction value.
Compare the stress set by the program shown in the flowchart with a predetermined value (for example, 0.8) to determine the air-fuel ratio feedback correction coefficient L when learning is almost completed.
It is determined whether the degree of variation in MD (degree of variation in base air-fuel ratio) is greater than or equal to a predetermined value.

Here, when the stress exceeds a predetermined value, it is determined that the learning result is inappropriate and an air-fuel ratio deviation has occurred, although the learning is almost completed, and the learning correction coefficient KBLRC φ is The process advances to step 92 in order to perform the learning again. In step 92, flags Fφ, F[0,0] to F[3,3], FF[0,
0] to FF[16, 16] are all reset to zero, and since learning will be restarted from the beginning as described above, stress is also reset to zero.

In this way, if the degree of deviation of the air-fuel ratio feedback correction coefficient LMD from the reference value becomes larger than a predetermined value, relearning can be performed. When the fuel ratio suddenly changes, learning for each large operating range is performed again, so the air-fuel ratio can be quickly converged to the target air-fuel ratio.

[0069]

[Effects of the Invention] As explained above, according to the present invention, it is possible to perform detailed learning for each operating region that can respond to differences in operating conditions while ensuring a convergence speed of learning, and to perform highly accurate interpolation. Through calculation, the air-fuel ratio learning correction value can be changed smoothly in response to changes in operating conditions, and it is possible to prevent air-fuel ratio fluctuations from occurring when changing the corresponding operating region on the learning map and deteriorating exhaust properties. There is an effect.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of the present invention.

FIG. 2 is a system schematic diagram showing an embodiment of the present invention.

FIG. 3 is a flowchart showing air-fuel ratio feedback control.

FIG. 4 is a flowchart showing air-fuel ratio feedback control.

FIG. 5 is a flowchart showing air-fuel ratio learning control.

FIG. 6 is a flowchart showing air-fuel ratio learning control.

FIG. 7 is a flowchart showing air-fuel ratio learning control.

FIG. 8 is a flowchart showing air-fuel ratio learning control.

FIG. 9 is a flowchart showing correction and readout control of learning results.

FIG. 10 is a flowchart showing setting of fuel injection amount.

FIG. 11 is a flowchart showing contents related to learning iterative control.

FIG. 12 is a diagram showing the state of the learning map in the example.

FIG. 13 is a diagram showing how operating conditions are set in interpolation calculations.

FIG. 14 is a diagram showing characteristics of interpolation calculations.

FIG. 15 is a time chart for explaining problems with conventional learning control.

[Explanation of symbols]

1. Institution 6 Fuel injection valve 12 Control unit 13 Air flow meter 14 Crank angle sensor 15 Water temperature sensor 16 Oxygen sensor

Claims (2)

[Claims] 1. Engine operating condition detection means for detecting engine operating conditions including at least operating parameters related to the amount of air taken into the engine; A basic fuel supply amount setting means for setting the fuel supply amount, an air-fuel ratio detection means for detecting the air-fuel ratio of the engine intake air-fuel mixture, and a comparison between the air-fuel ratio detected by the air-fuel ratio detection means and the target air-fuel ratio. an air-fuel ratio feedback correction value setting means for setting an air-fuel ratio feedback correction value for correcting the basic fuel supply amount so that the actual air-fuel ratio approaches the target air-fuel ratio; a storage means for rewritably storing an air-fuel ratio learning correction value for correcting the basic fuel supply amount for each operating region; and a storage means for learning a deviation of the air-fuel ratio feedback correction value from a target convergence value. The air-fuel ratio learning is performed by correcting and rewriting the air-fuel ratio learning correction value stored corresponding to the corresponding operating region in a direction that reduces the deviation, and the range of the corresponding operating region is gradually narrowed as learning progresses. reading out the air-fuel ratio learning correction value corresponding to the actual operating conditions from the storage means, with an interpolation calculation using the narrowest operating region learned by the air-fuel ratio learning means as a unit; and a fuel supply amount setting that sets a final fuel supply amount based on the basic fuel supply amount, the air-fuel ratio feedback correction value, and the air-fuel ratio learning correction value read by the correction value reading means. and fuel supply control means for driving and controlling the fuel supply means based on the fuel supply amount set by the fuel supply amount setting means. Device. 2. The interpolation calculation by the correction value reading means sets a substantially central value of the operating condition width of each operating region as the operating condition to which the air-fuel ratio learning correction value in each operating region corresponds, and calculates the actual operating condition. The air-fuel ratio learning control device for an internal combustion engine according to claim 1, wherein the air-fuel ratio learning control device for an internal combustion engine is configured to obtain an air-fuel ratio learning correction value corresponding to the air-fuel ratio learning correction value.